Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 1506 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.1143. Min Act: -0.0001

Data Index: 414883 (C4 (Web Text))

Max Activating Token Index: 556

Text #1

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.1112. Min Act: -0.0001

Data Index: 198701 (C4 (Web Text))

Max Activating Token Index: 126

Text #2

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.1104. Min Act: -0.0001

Data Index: 629351 (C4 (Web Text))

Max Activating Token Index: 29

Text #3

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.1071. Min Act: -0.0001

Data Index: 300850 (C4 (Web Text))

Max Activating Token Index: 712

Text #4

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.1127. Min Act: -0.0001

Data Index: 1125697 (C4 (Web Text))

Max Activating Token Index: 117

Text #5

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.1014. Min Act: -0.0001

Data Index: 1463117 (Python Code)

Max Activating Token Index: 46

Text #6

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.0981. Min Act: -0.0001

Data Index: 1196849 (C4 (Web Text))

Max Activating Token Index: 170

Text #7

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.0948. Min Act: -0.0001

Data Index: 968239 (C4 (Web Text))

Max Activating Token Index: 794

Text #8

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.0935. Min Act: -0.0001

Data Index: 1302039 (C4 (Web Text))

Max Activating Token Index: 732

Text #9

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.1012. Min Act: -0.0001

Data Index: 110860 (C4 (Web Text))

Max Activating Token Index: 861

Text #10

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.0936. Min Act: -0.0001

Data Index: 636758 (C4 (Web Text))

Max Activating Token Index: 986

Text #11

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.0904. Min Act: -0.0001

Data Index: 1126358 (C4 (Web Text))

Max Activating Token Index: 300

Text #12

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.0964. Min Act: -0.0001

Data Index: 633317 (C4 (Web Text))

Max Activating Token Index: 930

Text #13

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.1037. Min Act: -0.0001

Data Index: 49750 (C4 (Web Text))

Max Activating Token Index: 708

Text #14

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.0985. Min Act: -0.0001

Data Index: 1461306 (Python Code)

Max Activating Token Index: 421

Text #15

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.0942. Min Act: -0.0001

Data Index: 253629 (C4 (Web Text))

Max Activating Token Index: 949

Text #16

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.0931. Min Act: -0.0001

Data Index: 190014 (C4 (Web Text))

Max Activating Token Index: 795

Text #17

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.0977. Min Act: -0.0001

Data Index: 783627 (C4 (Web Text))

Max Activating Token Index: 950

Text #18

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.0870. Min Act: -0.0001

Data Index: 1411037 (Python Code)

Max Activating Token Index: 742

Text #19

Max Range: 0.1143. Min Range: -0.1143

Max Act: 0.0899. Min Act: -0.0001

Data Index: 599354 (C4 (Web Text))

Max Activating Token Index: 796